BOINC 6.6.11 is out. As stated in other threads this is a Development version. The change log is below.

6.6.11

- client: if a project is at max backoff for a resource, stop accumulating debt if it's at or around zero.

This prevents other projects from being driven unboundedly negative.

- client: if the number of overworked projects exceeds the number of device instances, clear debts; this indicates that an earlier client was buggy and produced bad debt values.

- client: new work-fetch policy:

1) if an instance is idle, get work from highest-debt project, even if it's overworked.

2) if resource has a shortfall, get work from highest-debt non-overworked project

3) if there's a fetchable non-overworked project with no runnable jobs, get from the highest-debt one
(each step is done first for GPU, then CPU)

Clause 3) is new.

It will cause the client to get jobs for as many projects as possible, even if there is no shortfall. This is necessary to make the notion of "overworked" meaningful (otherwise, any project with long jobs can become overworked). It also maintains as much variety as possible (like pre-6.6 clients).

Also (small bug fix) if a project is overworked for resource R, request work for R only in case 1).

- client: remove the "debt repair" mechanism added earlier today.

There are situations where multiple projects can legitimately have large negative LTD on a uniprocessor. Instead...

- client: add <zero_debts> option to cc_config.xml

- client: fix bug that can cause no GPU jobs to be scheduled even when a GPU is free

- client: tweak the LTD calculation so that the largest debt always tends towards zero

- client: print message if downloaded file has wrong size

- client: associate file xfer messages with a project; fixes #848

- client: more instances of showing project with message. Fixes #848

- client: fix work-fetch bug that caused infinite fetch; cleanup/reorganization of work fetch logic

- client: work-fetch bug fix: if we're fetching work for a starved project, it most have no runnable jobs for ANY resource.

- client: work-fetch bug fix: when setting requests in the shortfall case, don't request anything if project is backed off or overworked for the resource.

- scheduler: include driver version in the CUDA description string storing in the database;

- WINSCR: Fix the BOINC text displayed is the screensaver control panel applet

- WINSCR: Get rid of the 'Up to 5 second wait' for the data thread to terminate. Force terminate the thread and clean up the graphics app in a clean-up routine. Forcing the user to wait when they return to the system isn't a good idea.

- WINSCR: Fix screensaver issue where the screensaver locks up when the graphics application deadlocks for one reason or another.

- WINSCR: Clean-up some of the thread cleanup logic.

---

BOINC blog

So is there any testers of this new version?

---

From Siberia with love!

So is there any testers of this new version?

6.6.11 is running on several PCs since yesterday. In case there are problems I will post them in this thread.

Please do not post only problems, but also successful observations of 6.6.11. A lot of people talk about problems, but no one about success.

6.6.11 is taking work for one of my projects even if the project is suspended and has the "no new work" flag (seti@home beta cuda); if i abort the work will take again work even if it's not supposed to do so.
for gpugrid it's working fine, keeps my cache full (4 jobs for my quad/260 computer";
for MW is better than 6.5 now that people are struggling for MW wu's

---

@ Nowi

I've made a signature with the BOINC version currently running on my PCs.
As long as no older version is shown I encountered no severe problems with it.
Is this a reasonable way to do it?

/edit/ I improved formulation.

---

Please do not post only problems, but also successful observations of 6.6.11. A lot of people talk about problems, but no one about success.

Got it installed on 1 machine at the moment. Doesn't appear to have any problems, but I need to retest a few things that were broken in earlier 6.6 builds. Work-fetch issues usually take a couple of days of running to come out.

---

BOINC blog

Got it installed on 1 machine at the moment. Doesn't appear to have any problems, but I need to retest a few things that were broken in earlier 6.6 builds. Work-fetch issues usually take a couple of days of running to come out.

Make that 2 machines now. Also installed latest nVidia drivers (182.06). Checked a couple of things from prior fixes and they still seem to work. All looking good so far.

---

BOINC blog

It is still too soon for me to know about 6.6.11, but I have noticed work
fetch happening from the lowest LTD projects when there is sufficient work
available from higher LTD projects. To help restore LTD balance, work
fetch should be more heavily balanced toward the highest LTD tasks.

jm7

From the Dev list ... I do note that this may have been addressed in .11 as there are additional changes to fetch policy.

6.6.11 is taking work for one of my projects even if the project is suspended and has the "no new work" flag (seti@home beta cuda); if i abort the work will take again work even if it's not supposed to do so.
for gpugrid it's working fine, keeps my cache full (4 jobs for my quad/260 computer";
for MW is better than 6.5 now that people are struggling for MW wu's

I have the same problem:
---
26/02/2009 20:24:08 GPUGRID Sending scheduler request: Requested by user.
26/02/2009 20:24:08 GPUGRID Reporting 4 completed tasks, not requesting new tasks
26/02/2009 20:24:14 GPUGRID Scheduler request completed: got 1 new tasks
---
Random.
Also with Einstein@home.

I tried it on the system running MW and it would not fetch work reliably for me (or as reliably as 6.5.0 did), so I went back ...

And that was an adventure too ... it did not want to install 6.5.0 ... after 30 minutes migrating data I shot the installer and finally got it to install the older version ...

I sure wish they would get this un-kerfuflled ...

yeah, i've backed down to 6.5 also for the same reasons (no MW work).

---

I have also tested it, but I encountered problems with the work fetch of EAH-WU. Fall back to 6.5.0

Using 6.6.11 since two days (with WCG, superlink and GPUGRID) there are no problems so far.

---

NNW does not work for GPU projects. So if you have more than one attached, and you set one to NNW, you will still get work from it.

---

Reno, NV
Team: SETI.USA

6.6.11 for Linux (64 Bit) still doesn't get WUs from other projects than climateprediction. All other just say:

"Scheduler request commpleted: got 0 new tasks."

---

GPUGRID on 64 Bit Linux with GeForce 9400GT (512MB), AMD X2 4850e (2 * 2,5 GHz), 8 GB RAM

Please do not post only problems, but also successful observations of 6.6.11. A lot of people talk about problems, but no one about success.

No problems so far. I have it running on 4 machines.

---